AI RESEARCH
Multimodal Backdoor Attack on VLMs for Autonomous Driving via Graffiti and Cross-Lingual Triggers
arXiv CS.CV
•
ArXi:2604.04630v1 Announce Type: new Visual language model (VLM) is rapidly being integrated into safety-critical systems such as autonomous driving, making it an important attack surface for potential backdoor attacks. Existing backdoor attacks mainly rely on unimodal, explicit, and easily detectable triggers, making it difficult to construct both covert and stable attack channels in autonomous driving scenarios. GLA